Published on Web 06/28/2005
Studies of C-S Bond Cleavage Reactions of Re(V) Dithiolates: Synthesis, Reactivity, and Mechanism Ming Li, Arkady Ellern, and James H. Espenson* Contribution from the Ames Laboratory and Department of Chemistry, Iowa State UniVersity, Ames, Iowa 50011 Received March 25, 2005; E-mail:
[email protected] Abstract: A series of rhenium(V) complexes, [(X)(ReO)(dt)(PPh3)] and [(o-SC6H4PPh2)(ReO)(mtp)], were prepared to explore electronic effects on the C-S cleavage reaction that occurs upon reaction with PAr3 at ambient temperature [where X ) S(C6H4-p-Z) (Z ) OMe, Me, H, F, Cl), OPh, Cl, and SC2H5, and dt is the chelating dithiolate ligand derived from 2-(mercaptomethyl)thiophenol, 1,2-ethanedithiol, 1,3-propanedithiol, 1,3-butanedithiol, and 2,4-pentanedithiol]. The scope and selectivity of the C-S activation were examined. The C-S bond cleavage to form metallacyclic Re(V) complexes with a RetS core occurs only for the complexes with mtp and pdt frameworks and X ) SAr and SC2H5. The difference in reactivity is due to the different donating abilities of ancillary and dithiolate ligands, especially their π-donating ability, which plays a critical role in C-S activation. The kinetics of the C-S activation process was determined; nucleophilic attack of PPh3 on the oxo group of the ReVO core appears to be the rate-controlling step. The reaction is accelerated by electron-poor ArS ligands, but is unaffected by the substituents on phosphines. A detailed mechanistic study is presented. The results represent a rare example of migration of alkanethiolate leading to the formation of alkylthiolato complexes.
Introduction
The chemical reactivities of oxorhenium(V) complexes containing a chelated dithiolate (dt) ligand1 pose dramatic contrasts depending on whether the “bystander” ligand is PhS or Me.2 These chemical equations show the reactions with PPh3/ PAr3 that occur readily at ambient temperature, illustrating the sharp distinction between the two.
terminal Re-S bond is made at the expense of C-S-Re bonds, and a C-Re bond is formed. The Me-Re analogue, on the other hand, undergoes only ligand substitution in sequential steps, eq 3.3 This article presents new data for the reactions represented by eqs 1 and 2 through variations of the constitution of the rhenium reagents, as shown in the structural formulas for 1, 3, 5, 6, and 7 in Chart 1. The kinetics of these reactions was studied to determine the chemical mechanism. Additionally, attention has been devoted to understanding the contrasting behavior of the PhS-Re and Me-Re compounds in light of further findings concerning [XReO(dt)(PPh3)], 5 and 6, where X ) Cl, PhO, and EtS in addition to PhS. Results
Equations 1 and 2 represent the results of the evidently complex process that is the principal focus of this article. In the course of these reactions, the Re-O bond is broken, a (1) Abbreviations: mtpH2, 2-(mercaptomethyl)thiophenol; edtH2, 1,2-ethanedithiol; pdtH2, 1,3-propanedithiol; pdtMeH2, 1,3-butanedithiol; pdtMe2H2, 2,4pentanedithiol. Stand-alone formulas for the ligands in complexes are written as mtp, edt, etc., without charge being shown. (2) Li, M.; Ellern, A.; Espenson, J. H. Angew. Chem., Int. Ed. 2004, 43, 58375839. 10436
9
J. AM. CHEM. SOC. 2005, 127, 10436-10447
Synthesis of Rhenium(V) Complexes. The most efficient route to monomeric oxorhenium(V) complexes is by reaction of the more readily prepared dimers.4 In this study, compounds 1 and 5b were prepared by converting the recently reported,5 brown-colored dimers, {PhSReO(dt)}2, to the green monomers, [PhSReO(dt)(PPh3)], by reaction with a controlled quantity of PPh3, namely a 1:2 ratio of dimer to PPh3. The products were characterized by 1H and 31P NMR spectroscopy and by elemental analysis. The experimental details for these reactions, shown in Scheme 1, are described in the Experimental Section. These reactions proceed in high yield for dt ) mtp and edt (3) Lahti, D. W.; Espenson, J. H. J. Am. Chem. Soc. 2001, 123, 6014-6024. (4) Jacob, J.; Lente, G.; Guzei, I. A.; Espenson, J. H. Inorg. Chem 1999, 38, 3762-3763. Lente, G.; Guzei, I. A.; Espenson, J. H. Inorg. Chem 2000, 39, 1311-1319. (5) Li, M.; Ellern, A.; Espenson, J. H. Inorg. Chem 2005, 44, 3690-3699. 10.1021/ja0519334 CCC: $30.25 © 2005 American Chemical Society
C−S Bond Cleavage Reactions of Re(V) Dithiolates
ARTICLES
ature syntheses of [ClReO(dt)(PPh3)] with dt ) edt (5a) and mtp (6a) were developed. The method proved to be useful for preparing precursors to a variety of complexes with different ancillary ligand environments. Reactions of [Cl3ReO(PPh3)2] and 1.0 equiv of edt or mtp in toluene afford products 5a and 6a, respectively.
[Cl3ReO(PPh3)2] + dtH2 f [ClReO(dt)(PPh3)] + PPh3 + 2HCl (dt ) edt, mtp) (4)
Figure 1. ORTEP diagram of 1a with thermal ellipsoids drawn to 50% probability. Chart 1
Scheme 1
because monomerization proceeds much more readily than the C-S bond cleavage reactions. With dt ) pdt, pdtMe, and pdtMe2, however, mixtures of monomers were obtained due to stereoisomers and the competing formation of C-S cleavage products; no attempts were made to isolate the monomers. The structure of 1a in the solid state was determined by X-ray crystallography. An ORTEP diagram is shown in Figure 1, and selected bond lengths and angles are listed in Table 1. The structure is very close to that of its analogue, [MeReO(mtp)(PPh3)], 9. The dithiolate ligand in 1a is bound asymmetrically with Re-S1 ) 233.46(10) pm and Re-S2 ) 228.44(10) pm; the Re-O distance is 169.3(3) pm. These distances are comparable to those found in 9, for which the three respective bond distances are 234.0, 227.54, and 169.6 pm.6 The monomerization reaction (Scheme 1) was used to prepare the ReVO complexes with various phosphines. Room-temper(6) Jacob, J.; Guzei, I. A.; Espenson, J. H. Inorg. Chem. 1999, 38, 10401041.
The 1H and 31P NMR spectra and the elemental analyses are consistent with the given formulas. Further evidence comes from the single-crystal X-ray structures. The ORTEP diagrams of the molecular structures are shown in Figure 2. These squarepyramidal molecules resemble 1a, with Re-Cl distances of 234.29(19) and 232.62(18) pm for 5a and 6a, respectively, which are similar to those in related Re(V) dithiolate complexes.7 The chloride ion is positioned trans to the benzylic sulfur atom. Attempts to synthesize [ClReO(dt)(PPh3)] (dtH2 ) pdtH2, 1,2-benzenedimethanethiol, and 1,4-butanedithiol) by a similar procedure were unsuccessful. With 5a and 6a in hand, substitution reactions with ancillary ligands were examined. Treatment of 6a with 1.0 equiv of LiSC6H4-p-X (X ) OMe, Me, F, Cl) gave 1a-e as green solids in 51-68% yields (Scheme 2). The 1H NMR spectra of these complexes display very similar resonance patterns; their 31P NMR spectra show a singlet at 18.7 ( 0.1 ppm. Treatment of 6a with 1.0 equiv of the chelating ligand lithium 2-(diphenylphosphino)benzenethiolate in benzene resulted in displacement of PPh3 and formation of 7 (Scheme 2). Its 31P NMR spectrum exhibits a singlet at δ 55.9 ppm. The structure of 7 was confirmed by X-ray diffraction (Figure 3); selected distances and angles are given in Table 1. The coordination geometry about rhenium is square-pyramidal, with a Re-O length of 168.6(5) pm and Re-S bond lengths in the range of 227.89(18)-235.1(2) pm. The aryl ring of the chelating ligand o-SC6H4PPh2 is nearly coplanar with the P1-Re-S3 plane: the dihedral angle is 13.9°. The distances between rhenium and its donor atoms are only slightly affected, as can be seen from a comparison of the structures of 1a and 7. Addition of 1.0 equiv of LiSC2H5 to 6a followed by workup afforded 6b as a green solid. The 1H NMR spectrum established the presence of an ethyl group in which the CH3 protons appeared at δ 1.45 and the chemically inequivalent CH2 protons appeared as doublets of quintets at δ 3.86 and 3.69. A suitable crystal was obtained by vapor diffusion of hexane into a CH2Cl2 solution of the complex. The significant bond lengths and angles are listed in Table 1; an ORTEP diagram is shown in Figure 3. Treatment of 6a with 1.0 equiv of LiOPh in anaerobic THF led to the desired product (Scheme 2). The salient feature in the 1H NMR spectrum of 6c is the appearance of the methylene protons at δ 4.77 and 3.37, upshifted in comparison with 6a; the 31P resonance shifts from δ 23.0 (6a) to 32.5 ppm (6c). In a similar manner, treatment of 5a with LiSC2H5 at room temperature gave [EtSReO(edt)(PPh3)], 5c, after 3 h. (7) Gangopadhyay, J.; Sengupta, S.; Bhattacharyya, S.; Chakraborty, I.; Chakravorty, A. Inorg. Chem 2002, 41, 2616-2622. Dirghangi, B. K.; Menon, M.; Pramanik, A.; Chakravorty, A. Inorg. Chem. 1997, 36, 10951101. J. AM. CHEM. SOC.
9
VOL. 127, NO. 29, 2005 10437
Li et al.
ARTICLES Table 1. Selected Bond Lengths (pm) and Angles (deg) 1a
5ab
6a
6b
7
9c
Re-O Re-SPha Re-SBna Re-Xa Re-P
169.3(3) 233.46(10) 228.44(10) 230.09(10) 249.81(10)
169.3(5) 228.8(2) 225.30(19) 234.29(19) 247.99(17)
167.5(5) 231.7(2) 223.73(19) 232.62(18) 248.68(18)
168.8(4) 233.0(2) 228.06(18) 2.270(2) 247.86(17)
168.6(5) 235.1(2) 227.89(18) 230.79(18) 243.63(18)
169.6(5) 234.0(2) 227.54(19) 213.1(7) 244.68(19)
O-Re-SPha O-Re-SBna O-Re-P O-Re-X
105.78(10) 119.21(10) 92.10(10) 113.99(10)
108.63(17) 113.7(2) 99.47(17) 111.9(2)
104.95(19) 114.3(2) 92.87(19) 120.9(2)
105.48(17) 117.9(2) 92.11(17) 115.2(2)
106.15 (19) 115.03(18) 98.96(18) 113.68(18)
109.06(19) 116.14(19) 100.13(18) 112.1(3)
aS Ph is bound to the phenyl of mtp; SBn is the benzylic S, bound to the methylene group of mtp; X ) Cl for 6a and 5a; X ) S for 1a, 6b, and 7; and X ) C for 9. b Both SPh and SBn refer to the edt ligand. c Data from one of two independent molecules; ref 6.
low yield after column chromatography. Examination of the
Figure 2. ORTEP diagrams for 5a and 6a with thermal ellipsoids drawn to 50% probability.
Reactions of the mtp Complexes 1a-e with PPh3. As reported previously, treatment of 1a with excess PPh3 leads to the metalacyclic Re(V) complex 2a, eq 1. Similarly, C-S bond cleavage reactions also occur for [XC6H4SReO(mtp)(PPh3)] (1b-e, X ) OMe, Me, F, Cl). These reactions were confirmed by 1H and 31P NMR spectra. Products 2b-e show NMR patterns quite similar to that of 2a. The 1H NMR resonances of the methylene group of the mtp ligand of each product are downshifted in comparison with those of the corresponding reactants; the 31P NMR spectra exhibit resonances around δ 35.0. Treatment of 6b with excess PPh3 in benzene at room temperature allowed the isolation of 8 as a red-brown solid in 10438 J. AM. CHEM. SOC.
9
VOL. 127, NO. 29, 2005
entire reaction mixture by 1H NMR in C6D6 revealed that the conversion of 6b to 8 proceeded quantitatively, although the isolated yield was lower owing to decomposition on the chromatographic column. The 1H and 31P NMR spectra of 8 are consistent with the C-S bond rupture at the position trans to the ethanethiolate ligand. In particular, the δ 4.92 doublets of doublets (4JP-H ) 1.2 and 2JH-H ) 12 Hz) and the δ 3.35 doublet (2JH-H ) 12 Hz), assigned as the methylene protons of the benzylic moiety in the 1H NMR spectrum of 6b, are replaced by δ 6.67 (3JP-H ) 3.2 and 2JH-H ) 16.6 Hz) and δ 4.09 (3JP-H ) 9.6 and 2JH-H ) 16.6 Hz) doublets of doublets. Accordingly, the chemical shifts of the ethyl protons are slightly changed, although two sets of multiplets of the methylene protons of the ethyl group are collapsed to one set. Reactions of 6a, 6c, and 7 with PPh3 were monitored by 1H and 31P NMR spectra in C6D6. Neither a C-S cleaving reaction nor a triphenylphosphine oxide adduct of rhenium was observed, even upon prolonged reaction time or heating. It should be mentioned that the phosphorus resonance of 6a, upon addition of excess PPh3, broadens without changing its chemical shift, presumably due to PPh3 exchange. Reactions of pdt Complexes with PPh3. In reactions of the Re(V) complexes with an mtp ligand, 1a-e, C-S bond cleavage occurred only at the benzylic moiety. To learn whether this is intrinsic or due to the mtp ligand, complexes with pdt were prepared. Treatment of 3a with PPh3 yields the thoroughly characterized C-S cleavage product 4a. Its formation does not allow a distinction to be made as to the position of C-S bond scission, owing to the symmetry of the pdt ligand. Thus we
examined the reactions of asymmetric pdt isomers with PPh3. Reaction of 1,3-butanedithiol (pdtMe) with {(ReO)(SPh)3}25 in a ratio of 2:1 should, in principle, afford the three dimeric stereoisomers shown in Chart 2.
C−S Bond Cleavage Reactions of Re(V) Dithiolates
ARTICLES
Scheme 2
Figure 4. ORTEP diagram for 4b with thermal ellipsoids drawn to 50% probability. Selected distances (pm) and bonding angles (°): Re(1)-S(3), 208.99(10); Re(1)-C(4), 214.6(4); Re(1)-S(1), 226.88(9); Re(1)-S(2), 228.57(9); Re(1)-P(1), 244.00(8); S(3)-Re(1)-C(4), 107.75(11); S(3)Re(1)-S(1), 117.55(4); C(4)-Re(1)-S(1), 79.43(10); S(3)-Re(1)-S(2); 112.97(4); C(4)-Re(1)-S(2), 88.01(10); S(3)-Re(1)-P(1), 101.75(4); S(1)-Re(1)-P(1), 84.27(3); S(2)-Re(1)-P(1), 84.27(3). Chart 2
Figure 3. ORTEP diagrams for 6b and 7 with thermal ellipsoids drawn to 50% probability.
According to the 1H NMR spectra of the dimers used in the C-S cleaving reaction, the ratio of methyl groups pointing toward the RetO bond to those pointing away from that bond is 3.37:1. Two well-resolved sets of proton and phosphorus resonances of unequal intensity were observed in the NMR spectra of the C-S cleaved products in C6D6. The proton resonances were correlated by a 2D COSY spectrum (Figure S1, Supporting Information); the phosphorus resonances appear at δ 31.8 and 32.3 for 4b and 4c, respectively. Integration of
the CH3 resonances affords a ratio of 3.26:1, which is in excellent agreement with that of the reactants. The identities of isomers 4b (major) and 4c were assigned on the basis of 2D COSY and HMQC C-H correlation spectra.
The structure of 4b was confirmed by X-ray diffraction; the ORTEP diagram and significant bond lengths and angles are shown in Figure 4. The methyl group lies on the same side of J. AM. CHEM. SOC.
9
VOL. 127, NO. 29, 2005 10439
Li et al.
ARTICLES
Figure 6. Variation of 31P NMR spectra with time in CD2Cl2 for monomerization of the dimer with the pdtMe ligand (6.5 mM) with PPh3 (77 mM): 15 min (bottom), 2 h (middle), and 4 h (top).
Figure 5. 31P NMR spectra of monomerization of the dimers with pdt (A) and mtp (B) ligands with PPh3 in CD2Cl2.
the molecule as the sulfide-rhenium bond; the RetO distance is 208.99(10) pm. The Re-C bond distance in 4b is 214.6(4) pm, comparable to those found in 4a and other related Re(V)-S complexes.8 To learn whether C-S bond scission is regioselective, the monomerization of dimers with mtp, pdt, and pdtMe ligands was monitored in a 1:2+ ratio of dimer to PPh3 by 1H and 31P NMR spectroscopy in CD2Cl2. For the dimer with a pdt ligand, one intensified resonance at 17.7 ppm assignable to 3a was observed in the 31P NMR spectra in Figure 5A, in addition to 4a at 31.2 ppm, free PPh3 at -4.8 ppm, and Ph3PO at 28.1 ppm. Unexpectedly, two monomeric isomers with the mtp ligand were detected as shown in Figure 5B. The resonance due to 1a appears at 18.6 ppm and that due to 1a′ at 17.0 ppm; moreover, isomer 1a′ was converted rapidly to 1a by PPh3 catalysis.3
isomers 3b/3b′ and 3c/3c′ disappeared to form only 4b (δ 30.9 ppm) and 4c (δ 31.5 ppm), respectively. The results show clearly that the C-S cleavage is not regioselective.
The difference in monomerization with asymmetric mtp and pdtMe can perhaps be attributed to the coordination abilities of the thiolato donors. In mtp, the sulfur atom directly connected to the phenyl ring coordinates to the rhenium center less strongly than the sulfur atom connected to the methylene group. This favors formation of the thermodynamic isomer 1a. Both sulfur atoms in pdt have the same coordination abilities, and this system lacks a driving force for isomer interconversion. Reactions of edt Complexes with PPh3. C-S bond activation occurs selectively at the six-membered rings with the mtp and pdt ligands as shown above, which results in the formation of stable five-membered rings. To extend the reactions further and to examine their selectivity, the geometrically constrained complex [EtSReO(edt)(PPh3)], 5c, was used for reaction with PPh3. The process was investigated by 1H and 31P NMR spectroscopy. As it happens, no reaction was observed.
In monomerization of the dimer with the pdtMe ligand, two pairs of resonances with roughly equal intensity were observed; those at 17.4 and 16.8 ppm were assigned to 3b and 3b′ and those at 17.1 and 15.6 ppm to 3c and 3c′. In contrast to the case with the mtp ligand, no interconversion between 3b and 3b′, or between 3c and 3c′, was observed, even in a 1:12 ratio of the dimer to PPh3 (Figure 6). As the reaction proceeded, the (8) Jacob, J.; Guzei, I. A.; Espenson, J. H. Inorg. Chem. 1999, 38, 32663267. 10440 J. AM. CHEM. SOC.
9
VOL. 127, NO. 29, 2005
Kinetics of the C-S Cleavage Reactions. These reactions were monitored by their UV-visible spectroscopic changes. The
C−S Bond Cleavage Reactions of Re(V) Dithiolates
Figure 7. (Top) Changes in the UV-visible spectrum observed during the reaction of 1a with PPh3. The respective initial concentrations of 1a and PPh3 were 1.26 × 10-4 and 0.329 mol L-1. The spectra were recorded at 25 °C at 10.0 min intervals. (Bottom) Plot of kobs against the concentration of PPh3.
time course of the reaction between 1a and PPh3 is depicted in Figure 7; the tight isosbestic points show that the reaction progresses to the product without intermediates building up to concentrations that contribute appreciably to the spectral changes. The final spectrum is identical with that of 2a measured separately. Global fitting of the spectra (350-650 nm) with SpecFit9 establishes that the reaction follows pseudo-first-order kinetics at all of these wavelengths under these conditions. Values of the first-order rate constants proved to be independent of the initial concentration of the Re(V) complexes, with kobs linearly dependent on [PAr3] (Figure 7 and Supporting Information). All reactions adhere to the second-order rate law,
-
d[Re(V)] ) k2[Re(V)][PAr3] dt
Rate constants are listed in Table 2. To explore the independent effects of the phosphine coordinated to 1a and of the external phosphine that attacks it, several variations were carried out. The reaction between 1a and P(C6H4-p-OMe)3 gave k2 ) 2.37(4) × 10-4 L mol-1 s-1, whereas that between an analogue of 1a, [PhSReO(mtp)P(C6H4-F-OMe)3], and P(C6H4-p-OMe)3 has k2 ) 2.23(3) × 10-4 L mol-1 s-1. Furthermore, the reaction between 1a and P(C6H4-p-CF3)3 has k2 ) 2.78(6) × 10-4 L (9) Binstead, R. A.; Zuberbuhler, A. D.; Jung, B. SPECFIT, Global analysis system; Spectrum Software Associates, 1993-2003.
ARTICLES
mol-1 s-1, comparable to that between 1a and PPh3. These data demonstrate that the rate constants are insensitive to the substituents (Y) in P(Ph-p-Y)3 (Y ) H, MeO, CF3). On the other hand, the reactions between [(C6H4-p-X)SReO(mtp)(PPh3)] (X ) MeO, Me, H, F, Cl) and PPh3 are accelerated by an electron-withdrawing group, whereas an electron-donating group has the opposite effect. A plot of log k2 vs the Hammett substituent constant (σ) is presented in Figure 8; analysis of the data, excluding the point for MeO, gives a slope F ) 0.66. Clearly, the point for X ) MeO lies off the line. This substituent is the most strongly electron-withdrawing. Aside from noting the fact, however, we offer no explanation for the unusual behavior. The rate constants also vary with changes of dithiolate ligands. In a series of complexes with various pdt ligands, the rates decrease with increasing numbers of methyl groups on the pdt framework. A plot of log k2 vs numbers of the methyl group exhibits a good linear correlation (R ) 0.97, Figure S8, Supporting Information). Infrared Spectra and Electrochemistry. The infrared spectra of the complexes (KBr disk) show the characteristic of RetO band, ranging from 978 to 947 cm-1. These values are within the range for the RetO stretching frequencies reported in the literature.7,10 Cyclic voltammetry was used to investigate the electrochemical behavior of the complexes, with results as summarized in Table 3. Compound 7 exhibited a reversible one-electron reduction; the others displayed quasi-reversible or irreversible voltammograms (see Supporting Information). Comparison of the reduction potentials in a series of mtp complexes indicates that the potentials increase in the order OPh < Me < PhS ≈ EtS ≈ Cl ≈ chelating ligand, o-SC6H4PPh2. This trend shows that the potentials are sensitive to the nature of the ancillary ligand. Compound 5b, with a five-membered ring, is difficult to reduce. Discussion
Electronic Factors. It has been well-documented that the ReVO moiety undergoes oxygen atom transfer upon reduction of Re(V) to Re(III), the transfer process being initiated by nucleophilic substrate attack on the π* ReO orbital.11,12 A difference in the electron-donating abilities of the ligands is expected to influence the reactivity dramatically. Electronwithdrawing ArS ligands facilitate nucleophilic attack, opposite to the effect of electron-rich ligands. For the reactions between 1a and a series of para-substituted phosphines, P(Ph-p-Y)3 (Y ) H, MeO, CF3), the rate reaction constants are nearly but not precisely the same (Table 3). The kinetic data raise two issues. First, note the identity (within three standard deviations) of the values of k2 for reactions when P(C6H4-p-OMe)3 is the attacking phosphine. Phosphine exchange between 1a and free phosphine occurs3 prior to C-S bond cleavage, as evidenced by the 31P NMR spectra. The C-S cleavage step with PAr3 is thus a single reaction between [PhSReO(mtp)P(C6H4-p-OMe)3] and P(C6H4-p-OMe)3. (10) Parachristou, M.; Pirmettis, I. C.; Tsoukalas, C.; Papagiannopoulou, D.; Raptopoulou, C.; Terzis, A.; Stassinopoulou, C. I.; Chiotellis, E.; Pelecanou, M.; Papadopoulos, M. Inorg. Chem 2003, 42, 5778-5784. (11) Holm, R. H. Chem. ReV. 1987, 87, 1401-1449. Mayer, J. M. AdV. Trans. Met. Coord. Chem. 1996, 1, 105-157. (12) Nugent, W. A.; Mayer, J. M. Metal-Ligand Multiple Bonds; John Wiley & Sons: New York, 1988. J. AM. CHEM. SOC.
9
VOL. 127, NO. 29, 2005 10441
Li et al.
ARTICLES Table 2. Summary of the Rate Constants for Reactions between Oxorhenium(V) Dithiolates and Triarylphosphines in C6H6 at 25 °C [PhSReO(mtp)(PPh3)] + P(Ar-p-Y)3 X
10-4
2.77(4) × 2.37(4) × 10-4 2.78(6) × 10-4 2.23(3)× 10-4 a 2.69(7)× 10-4 b
H MeO CF3
a
[(Ar-p-X)SReO(mtp)(PPh3)] + PPh3
k2 (L mol-1 s-1)
Y
MeO Me H F Cl
10-4
2.70(7) × 2.18(4) × 10-4 2.77(4) × 10-4 3.31(7) × 10-4 3.96(6) × 10-4
dt
k2 (L mol-1 s-1)
pdt pdtMe pdtMe2
3.61(2) × 10-3 2.27(4) × 10-3 7.7(5) × 10-4
For [PhSReO(mtp)P(C6H4-p-OMe)3] and P(C6H4-p-OMe)3. b For [EtSReO(mtp)(PPh3)] and PPh3.
Figure 8. Plot of log k2 against the Hammett constant σ for reactions between [(C6H4-p-XS)ReO(mtp)(PPh3)] and PPh3. Table 3. Electrochemical Data for Reduction of the Oxorhenium(V) Complexesa complex
Ep,c (V)
complex
Ep,c (V)
6a 7 1a 6b
-0.84 -0.90 -0.95 -1.04
3a 9 5b 6c
-1.10 -1.23 (-1.25)b (-1.49) b
a
[PhSReO(dt)(PPh3)] + PPh3
k2 (L mol-1 s-1)
Potentials in CH2Cl2 referenced vs SCE. b Irreversible.
The second issue is more troubling. The inductive effects of the two phosphines in the transition state appear nearly to cancel one another. It is reasonable to assert that a key step is the attack of phosphine at the oxo group of 1a. Studies have amply verified this mode of attack on metal-oxo groups.13 Given that, the phosphine attacking 1a acts as a nucleophile, and for it the rate constants for P(C6H4X)3 can reasonably be expected to follow the sequence X ) MeO > H > CF3. Indeed, a strong inductive effect can be expected, because of data pertaining to reaction 1, for which a Hammett correlation gave the reaction constant F ) -0.70.1414
[MeReVIIO2(mtp)] + PAr3 f [MeReVO(mtp)(OPAr3)] (5) On the other hand, the ring substituent on the phosphine already coordinated to the rhenium atom should exert just the (13) Seymore, S. B.; Brown, S. N. Inorg. Chem 2000, 39, 325-332. Smith, P. D.; Millar, A. J.; Young, C. G.; Ghosh, A.; Basu, P. J. Am. Chem. Soc. 2000, 122, 9298-9299. Nemykin, V. N.; Davie, S. R.; Mondal, S.; Rubie, N.; Kirk, M. L.; Somogyi, A.; Basu, P. J. Am. Chem. Soc. 2002, 124, 756757. Basu, P.; Nemykin, V. N.; Sengar, R. S. Inorg. Chem 2003, 42, 74897501. (14) Wang, Y.; Espenson, J. H. Inorg. Chem 2002, 41, 2266-2274. 10442 J. AM. CHEM. SOC.
9
VOL. 127, NO. 29, 2005
opposite inductive effect. It appears that the opposing inductive effects nearly cancel one another; compare these values of k2/ 10-4 L mol-1 s-1: 2.37 (MeO), 2.77 (H), and 2.78 (CF3). This is not unreasonable, given that the two phosphines exert their opposite inductive effects in the same transition state. Mechanism. To probe the mechanistic features of the C-S bond-cleaving process, the reaction between 1a and PPh3 was examined in detail by 1H and 31P NMR spectra in C6D6 on the basis of its simplified 1H NMR spectroscopy in the aliphatic region. The variable spectra with time are shown in Figure 9. Three well-resolved pairs of resonances are observed in 1H NMR spectrum; their correlation is evident in the 2D COSY spectrum. The protons (Hs and Hs′) are present as the doublet peaks at δ 4.70 and 3.43 of 1a, which then disappeared as the reaction progressed. The δ 6.29 and 4.11 (Hp and Hp′) resonances were assigned to 2a, which builds up at the expense of 1a. The phosphorus signals are found at δ 19.0 for 1a and at δ 34.8 for 2a; a singlet at δ 25.9 ppm is due to Ph3PO, as confirmed by an authentic sample. Two doublets in the 1H spectrum at δ 4.95 and 3.71 (HI and HI′) were assigned to an intermediate, which increased and then disappeared. At most it amounted to ca.12%. No obvious Re-31P resonance appeared in the 31P spectra, although a minor transient resonance near δ 35 (the Re-31P resonance of the product) cannot be discounted. Therefore, four possible intermediates have to be considered. A three-coordinate species A can be ruled out due to a splitting (∆δ ) 1.3) in chemical shifts of the methylene protons. It is possible but unreasonable that dimeric isomers B and C are present in solution. An analogue, {ReO(bdt)(SPh)}2 (bdt ) o-benzenedithiol), reacted with PPh3 very rapidly,5 suggesting that, in the presence of excess PPh3, it would be impossible to detect the dimers B and C by 1H NMR. It is also expected that the extent of dimer formation would be very small with phosphine present in such large excess. On the basis of the observed inequivalent protons of the methylene group due to the anisotropic effect of the RetS core,15 and the absence of H-P coupling, the coordinatively unsaturated intermediate D′ is being proposed. The chemical conclusions are, however, considerably more consistent with earlier studies of Re(V) chemistry if structure D, and not D′, is assigned to the intermediate, as shown in Chart 3. The data from this study and from earlier work allow us to assign a plausible multistep mechanism. As demonstrated experimentally, phosphine exchange occurs prior to the main reaction. Studies of [MeReO(dt)PR3] systems show that exchange occurs sequentially, via double turnstile rotation processes as shown in Scheme 3. Note that intermediate I increases to a concentration where it can easily be detected.3 (15) Martin, D.; Piera, C.; Mazzi, U.; Rossin, R.; Solans, X.; Font-Bardia, M.; Suades, J. J. Chem. Soc., Dalton Trans. 2003, 3041-3045. Papadopoulos, M. S.; Pirmettis, I. C.; Pelecanou, M.; Raptopoulou, C. P.; Terzis, A.; Stassinopoulou, C. I.; Chiotellis, E. Inorg. Chem. 1996, 35, 7377-7383.
C−S Bond Cleavage Reactions of Re(V) Dithiolates
Figure 9. Variation of 1H and
31P
ARTICLES
NMR spectra with time in C6D6 for reaction of 1a (9.4 mM) and PPh3 (0.16 mol L-1) at 25 °C.
Chart 3. Structural Formulas of Possible Intermediates
Scheme 3
Scheme 4. Suggested Mechanism of C-S Cleavage
The suggested route to C-S cleavage is presented in Scheme 4. Following phosphine exchange, PAr3 attacks nucleophilically at the oxygen atom of ReVO in the rate-controlling step, which
affords a phosphine oxide complex of Re(III). In the transition state, a [1,2] migration of an alkyl group to rhenium is proposed. Events following formation of the transition state are speculative, of course, but we suggest that weakly coordinating Ar3PO is lost first, yielding D. Finally, D isomerizes to the final product 2. This, too, occurs by a turnstile rotation sequence analogous to that in Scheme 3. Recently, Mayer and co-workers have reported the rhenium(V) complexes (HBpz3)ReO(R)Cl (HBpz3 ) hydrotris(1-pyrazolyl)borate, R ) aryl, ethyl) undergo photochemical metalto-oxo migrations to form rhenium(III) alkoxides.16 Analogous migrations to terminal imido or sulfide complexes have also been proposed.17
However, the reverse reaction, rearrangement of an alkoxide to an oxo-alkyl, is rare.18 In the present cases, migration of alkanethiolate to form the alkylthiolato complex, to our knowledge, is unprecedented. Reactivity. The complexes with the RetO core exhibit a wide range of reactivity toward PPh3. The compounds [XReO(dt)(PPh3)] (X ) SPh or SEt; dt ) mtp or pdt) react with PPh3 to induce C-S activation. At other end of the spectrum are the complexes which show no analogous reaction. These include [XReO(mtp)(PPh3)] (X ) OPh, Me, Cl, and o-SC6H4PPh2) and [YReO(edt)(PPh3)] (Y ) SPh and SEt). A few examples of ReVO and TcVO complexes possessing a sulfur donor set can be reduced by PPh3 to form deoxygenated, stable Re(III) and Tc(III) phosphine adducts instead.19 The electrophilicity of the multiply bonded ligand is in large part controlled by the energy and character of the LUMO.12,20 The Epc values, lying between -0.84 and -1.49 V (Table 3), (16) Brown, S. N.; Mayer, J. M. J. Am. Chem. Soc. 1994, 116, 2219-2220. Brown, S. N.; Mayer, J. M. Organometallics 1995, 14, 2951-2960. (17) Nugent, W. A.; Harlow, R. L. J. Am. Chem. Soc. 1980, 102, 1759-1760. Vivanco, M.; Ruiz, J.; Floriani, C.; Chiesi-Villa, A.; Rizzoli, C. Organometallics 1993, 12, 1802-1810. (18) Asselt, A. v.; Burger, B. J.; Gibson, V. C.; Bercaw, J. E. J. Am. Chem. Soc. 1986, 108, 5347-5349. Nelson, J. E.; Parkin, G.; Bercaw, J. E. Organometallics 1992, 11, 2181-2189. Tahmassebi, S. K.; Conry, R. R.; Mayer, J. M. J. Am. Chem. Soc. 1993, 115, 7553-7554. (19) Dilworth, J. R.; Neaves, B. D.; Hutchinson, J. R.; Zubieta, J. A. Inorg. Chim. Acta 1982, 65, L223-L224. Davison, A.; Vries, N. D.; Dewan, J.; Jones, A. G. Inorg. Chim. Acta 1986, 120, L15-L16. Bolzati, C.; Refosco, F.; Tisato, F.; Bandoli, G.; Dolmella, A. Inorg. Chim. Acta 1992, 201, 7-10. Vries, N. D.; Jones, A. G.; Davison, A. Inorg. Chem. 1989, 28, 37283734. J. AM. CHEM. SOC.
9
VOL. 127, NO. 29, 2005 10443
Li et al.
ARTICLES Table 4. Summary of Crystal Data for Complexes 1a, 5a, 6a, 6b, and 4b
formula fw, g cryst syst space group a, Å b, Å c, Å R, deg β, deg γ, deg V, Å3 Z D(calcd), Mg/m3 abs coeff, mm-1 GOF R1,a wR2b [I > 2σ(I)] a
1a
6a
5a
6b
7
4b
C31H26OPReS3‚0.5C7H8 774.0 monoclinic C2/c 33.844(6) 9.5051(16) 22.643(4) 90 121.374(3) 90 6219.1(18) 8 1.652 4.188 1.035 0.0299, 0.0692
C25H21OPClReS2‚1.5C6H6 771.32 monoclinic P2(1)/c 10.247(2) 30.973(6) 10.280(2) 90 100.432(3) 90 2181.6(5) 4 1.596 4.076 1.123 0.0502, 0.1062
C20H19OPClReS2 592.09 monoclinic C2/c 17.345(4) 10.982(2) 23.929(5) 90 110.753(3) 90 4262.4(16) 8 1.845 6.105 1.079 0.0447, 0.0960
C27H26OPReS3 679.83 triclinic P1h 9.272(3) 11.444(3) 12.909(4) 82.415(5) 80.987(4) 76.334(5) 1308.1(6) 2 1.726 4.964 0.953 0.0364, 0.0893
C25H20OPReS3 649.76 monoclinic P2(1)/n 10.870(3) 18.284(5) 12.674(3) 90 110.978(5) 90 2351.9(6) 4 1.835 5.517 1.090 0.0405, 0.1092
C28H28PReS3 677.85 monoclinic P2(1)/n 8.2037(14) 19.187(3) 17.022(3) 90 95.461(3) 90 2667.2(8) 4 1.688 4.866 1.028 0.0264, 0.0596
R1 ) ∑||Fo| - |Fc||/∑|Fo|. b wR2 ) {∑[w(Fo2 - Fc2)2]/∑[w(Fo2)2]}1/2.
provide a relative measure of the electrophilicity of the complexes.21 The trend illustrates the influence of ligand variation on electron density at RetO. The methyl ligand is a much better σ donor, and phenoxide is both a good σ donor and a much stronger π donor,22 as compared with PhS. The effect is to leave more electron density on Re, raising the LUMO energy and reducing the contribution of the 2p orbital of the oxygen atom to the LUMO. These factors are consistent with the observed lower electrophilic reactivity of those complexes. Replacement of the mtp ligand by an edt ligand makes complex 5b more electron rich and thus decreases its reactivity compared to that of 1a. This point needs to be emphasized. Reactivity is determined in the initial step, in which nucleophilic phosphine attacks the oxo group of [PhSReO(dt)(PPh3)]. It is not sufficient to argue that the ultimate product from the edt complex, with a four-membered chelate ring, will be less stable than that from mtp or pdt, with five-membered rings. That argument may well apply to the subsequent steps, but not to the rate-controlling step. This is the reason for emphasizing the first step. The issues surrounding the failure of [PhSReO(edt)PPh3], 5b, and [EtSReO(edt)PPh3], 5c, to react need to be addressed further. It is too facile simply to comment that the products, with fourmembered chelate rings, would be too unstable. Single-electron reduction of [PhSReO(pdt)PPh3], 3a, is only marginally easier than that of 5b. Further, the usage of single-electron Epc values in describing atom transfer chemistry may not be on secure grounds, although others have made use of it.21 The best comparison can be made between 3a and 5b because they both contain alkanedithiolato ligands. It seems probable, therefore, that the initial PPh3 attack leads to an intermediate that more resembles RedO‚‚‚PPh3 than ReIIIrOPPh3. The edt complex cannot pass through the transition state, however, because of ring strain. Thus, PPh3 is released and no net reaction occurs. The values of -Epc for [XReO(mtp)(PPh3)] (X ) SPh, SEt, Cl, and o-SC6H4P(Ph)2) fall into a small range, 0.84-1.04 V. (20) Meyer, T. J.; Huynh, M. H. V. Inorg. Chem 2003, 42, 8140-8160. Crevier, T. J.; Bennett, B. K.; Soper, J. D.; Bowman, J. A.; Dehestani, A.; Hrovat, D. A.; Lovell, S.; Kaminsky, W.; Mayer, J. M. J. Am. Chem. Soc. 2001, 123, 1059-1071. (21) Dehestani, A.; Kaminsky, W.; Mayer, J. M. Inorg. Chem 2003, 42, 605611. Bennett, B. K.; Crevier, T. J.; DuMez, D. D.; Matano, Y.; McNeil, W. S.; Mayer, J. M. J. Organomet. Chem. 1999, 591, 96-103. (22) Chisholm, M. H.; Davidson, E. R.; Huffman, J. C.; Quinlan, K. B. J. Am. Chem. Soc. 2001, 123, 9652-9664. 10444 J. AM. CHEM. SOC.
9
VOL. 127, NO. 29, 2005
Because 1a and 6b react with PPh3 to promote C-S bond cleavage, it can expected that 6a and 7 would undergo the same reaction, given that their similar Epc values are slightly more anodic than those for 1a and 6b. This not the case. The difference in reactivity is most likely due to the different π-donating ability of ancillary ligands X. The chloro ligand is a poor π-donor and is expected to interact weakly with Re d-π orbitals. Chelation in 7 inhibits rotation about the Re-S bond, thus significantly weakening the SfRe π-donation because the thiolate-metal interaction depends on the Re-S-C(X) angle and the Re-S-C1-C2(X) dihedral angle.23 The Re-S bonds in 1a and 6b rotate, which enhances the interaction that induces C-S cleavage. It is the SfRe π-donation and the formation of stable five-membered rings that provide the driving force for C-S activation. Conclusions
The preparation and thorough characterization of a series of oxorhenium(V) complexes that possess various types of ancillary ligands, chelate ligands, and dithiolates have allowed us to probe the influences of each of these variables on C-S activation. In particular, it has proved possible to focus on the nature of ligandfRe π-interactions. Spectroscopic, electrochemical, structural, and kinetic data reveal the critical role played by ligandfRe π-interactions in regioselective C-S bond cleavage reactions. Experimental Section All operations were carried out under an atmosphere of argon or nitrogen using standard Schlenk techniques unless other stated. Benzene, toluene, diethyl ether, THF, and hexane were distilled from sodium; dichloromethane was distilled from calcium hydride. Other reagents were used as received unless otherwise noted. Preparation of the dimers, {ReO(dt)(SPh)}2, where H2dt represents 2-(mercaptomethyl)thiophenol, 1,2-ethanedithiol, 1,3-propanedithiol, and 1,3-butanedithiol, has been reported elsewhere.5 2-(Diphenylphosphino)benzenethiol,24 [Cl3ReO(23) Davis, M. I.; Orville, A. M.; Neese, F.; Zaleski, J. M.; Lipscomb, J. D.; Solomon, E. I. J. Am. Chem. Soc. 2002, 124, 602-614. Fox, D. C.; Fiedler, A. T.; Halfen, H. L.; Brunold, T. C.; Halfen, J. A. J. Am. Chem. Soc. 2004, 126, 7627-7638. Solomon, E. I.; Szilagyi, R. K.; George, S. D.; Basumallick, L. Chem. ReV. 2004, 104, 419-458. Fiedler, A. T.; Halfen, H. L.; Halfen, J. A.; Brunold, T. C. J. Am. Chem. Soc. 2005, 127, 16751689. (24) Block, E.; Ofori-Okai, G.; Zubieta, J. J. Am. Chem. Soc. 1989, 111, 23272329.
C−S Bond Cleavage Reactions of Re(V) Dithiolates (PPh3)2],25 and 2,4-pentanedithiol (H2pdtMe2)26 were synthesized according literature procedures. Infrared spectra were collected with a Bruker IFS66V FT-IR spectrometer as KBr pellets. NMR spectra were obtained on a Bruker DRX-400 at 298 K, with 1H NMR chemical shifts referenced internally to the C6D6 solvent peak at δ 7.16 and the CD2Cl2 solvent peaks at δ 5.32; the 31P NMR chemical shifts are referenced externally to 85% aqueous H3PO4. UV-visible spectra were recorded on benzene solutions at 298 K in quartz cuvettes using a Shimadzu UV-3101PC spectrophotometer with a temperature-controlled unit. Elemental analyses were performed at Iowa State University’s Analysis Center with a PerkinElmer CHN/S analyzer, model 2400-II. Cyclic voltammograms were measured with a BAS-100 potentiostat using a three-compartment cell with an Ag/AgNO3 (0.1 mol L-1 in CH3CN) reference electrode, a platinum-disk working electrode, and a platinum-wire auxiliary electrode. The electrolyte solution was 0.1 mol L-1 tetra-n-butylammonium tetrafluoroborate in CH2Cl2. The solution in the working compartment was kept anaerobic under an atmosphere of N2. Potentials are reported relative to the SCE. [PhSReO(mtp)(PPh3)] (1a). A glass flask was charged with the dimer {PhSReO(mtp)}2 (187 mg, 0.20 mmol), toluene (40 mL), and PPh3 (105.7 mg, 0.40 mmol). The mixture was stirred at ambient temperature for 2 h, yielding a green solution which was then evaporated to dryness in vacuo. The residue was washed with diethyl ether to give pure 1a, 230 mg, 80%. Dark green crystals were obtained from CH2Cl2-hexanes. NMR: 1H (CD2Cl2), δ 7.79-7.73 (m, 6H), 7.56-7.54 (m, 2H), 7.50-7.46 (m, 9H), 7.39-7.36 (m, 2H), 7.29-7.22 (m, 3H), 7.18-7.14 (m, 2H), 4.95 (dd, J1 ) 12 Hz, J2 ) 1.6 Hz, 1H), 3.30 (d, J ) 12 Hz, 1H); 31P (CD2Cl2), δ 18.6. IR (KBr pellet): 946 cm-1 (s, νRetO). UV-visible (C6H6): λmax/nm (log ), 580 (2.70), 427(3.46), 369 (3.64). Anal. Found (calcd) for C31H26S3PORe‚0.5C7H8: C, 53.39 (53.54); H, 3.62 (3.91); S, 11.66 (12.43); P, 3.41 (4.00). [PhSReO(edt)(PPh3)] (5b). Yield: 69%. NMR: 1H (CD2Cl2), δ 7.80-7.75 (m, 6H), 7.51-7.48 (m, 9H), 7.47-7.45 (m, 2H), 7.347.31 (m, 2H), 7.22-7.19 (m, 1H), 3.79-3.68 (m, 2H), 2.62-2.51 (m, 2H); 31P (CD2Cl2), δ 21.7. IR (KBr pellet): 949 cm-1 (s, νRetO). UVvisible (C6H6): λmax/nm (log ), 620 (1.72), 531 (2.45), 411 (3.55), 327(sh) (3.62). Anal. Found (calcd) for C26H24S3PORe: C, 46.40 (46.90); H, 3.47 (3.63); S, 13.98 (14.45). [PhSReO(pdt)(PPh3)] (3a). In an NMR tube, the dimer {PhSReO(pdt)}2 (6.68 mg, 8.0 µmol) and PPh3 (4.31 mg, 16 µmol) were mixed in CD2Cl2. After about 30 min, the solution color changed from brown to green. NMR: 1H, δ 7.84-7.79 (m, 6H), 7.59-7.46 (m, 9H), 7.407.28 (m, 4H), 7.23-7.19 (m, 1H), 3.43-3.39 (m, 1H), 3.33-3.24 (m, 1H), 3.13-3.10 (m, 1H), 3.03-2.96 (m, 1H), 2.25-2.46 (m, 1H), 2.402.31 (m, 1H); 31P, δ 17.7. [ClReO(mtp)(PPh3)] (6a). The reaction was conducted in air. 2-(Mercaptomethyl)thiophenol (mtpH2, 187.6 mg, 1.2 mmol) in 20 mL of toluene was added dropwise to a yellow suspension of [Cl3ReO(PPh3)2] (1.0 g, 1.2 mmol) in 280 mL of toluene, generating a clear green solution after several minutes. The reaction mixture was stirred at ambient temperature for 30 min and then filtered. The solvent was removed by rotary evaporation to afford a solid, which was washed with diethyl ether to give the pure green product (675 mg, 97%). NMR: 1H (CD2Cl2), δ 7.70-7.65 (m, 6H), 7.58-7.43 (m, 9H), 7.347.31 (m, 3H), 7.23-7.19 (t, 1H), 5.14 (d, J ) 11.6 Hz, 1H), 3.50 (d, J ) 11.6 Hz, 1H); 31P (CD2Cl2), δ 23.0. IR (KBr pellet): 975 cm-1 (s, νRetO). UV-visible (C6H6): λmax/nm (log ), 649 (1.77), 437 (2.94), 362 (3.37). Anal. Found (calcd) for C25H21S3RePCl‚0.5C4H10O: C, 46.81 (46.91); H, 3.77 (3.79); S, 8.59 (9.27). [X-p-C6H4SReO(mtp)(PPh3)] (X ) OMe, Me, F, Cl). A solution of 6a (98 mg, 0.15 mmol) in 60 mL of benzene was transferred by cannula to a white suspension of LiSC6H4-p-X (from HSC6H4-p-X and (25) Chatt, J.; Rowe, A. J. Chem. Soc. 1962, 4019-4033. (26) Overberger, C. G.; Ferraro, J. J.; Orttung, F. W. J. Org. Chem 1961, 26, 3458-3460.
ARTICLES BuLi) in a 1:1 mmol ratio. The resulting mixture was stirred for 2 h and then filtered. The solvent was evaporated to yield the solid products. Pure products were obtained after the solids were washed with diethyl ether. X ) OMe (1b). Yield: 66%. NMR: 1H (CD2Cl2), δ 7.78-7.73 (m, 6H), 7.58-7.45 (m, 9H), 7.41-7.36 (m, 2H), 7.28-7.24 (m, 3H), 7.14-7.10 (m, 1H), 6.93-6.90 (m, 1H), 4.93 (dd, Hz, J1 ) 12 Hz, J2 ) 1.6, 1H), 3.38 (s, 3H), 3.29 (d, J1 ) 12 Hz, 1H); 31P (CD2Cl2), δ 18.8. IR (KBr pellet): 947 cm-1 (s, νRetO). UV-visible (C6H6): λmax/ nm (log ), 584 (2.72), 428 (3.43), 370 (3.59). Anal. Found (calcd) for C32H28S3PO2Re: C, 51.16 (50.71); H, 4.32 (3.72); S, 12.07 (12.69). X ) Me (1c). Yield: 65%. NMR: 1H (CD2Cl2), δ 7.78-7.73 (m, 6H), 7.57-7.46 (m, 9H), 7.37-7.35 (m, 2H), 7.27-7.24 (m, 3H), 7.217.19 (m, 2H), 7.13-7.10 (m, 1H), 4.94 (dd, J1 ) 12 Hz, J2 ) 1.6 Hz, 1H), 3.29 (d, J1 ) 12 Hz, 1H), 2.43 (s, 3H). 31P NMR (CD2Cl2), δ 18.6. IR (KBr pellet): 950 cm-1 (s, νRetO). UV-visible (C6H6): λmax/ nm (log ), 581 (2.58), 428 (3.37), 370 (3.52). Anal. Found (calcd) for C32H28S3PORe: C, 51.67 (51.80); H, 4.26 (3.80); S, 12.36 (12.97). X ) F (1d). Yield: 68%. NMR: 1H (CD2Cl2) δ 7.77-7.73 (m, 6H), 7.57-7.45 (m, 9H), 7.37-7.35 (m, 2H), 7.27-7.25 (m, 3H), 7.157.10 (m, 1H), 7.09-7.06 (m, 2H), 4.96 (dd, J1 ) 12 Hz, J2 ) 1.6 Hz, 1H), 3.30 (d, J1 ) 12 Hz, 1H); 31P NMR (CD2Cl2), δ 18.8. IR (KBr pellet): 953 cm-1 (s, νRetO). UV-visible (C6H6): λmax/nm (log ), 578 (2.75), 430 (3.43), 370 (3.63). Anal. Found (calcd) for C31H25S3POFRe: C, 50.58 (49.92); H, 4.02 (3.38); S, 11.70 (12.90). X ) Cl (1e). Yield: 51%. NMR: 1H (CD2Cl2), δ 7.77-7.72 (m, 6H), 7.57-7.46 (m, 9H), 7.45-7.43 (m, 2H), 7.36-7.33 (m, 2H), 7.287.25 (m, 3H), 7.15-7.12 (m, 1H), 4.96 (dd, J1 ) 12 Hz, J2 ) 1.6 Hz, 1H), 3.30 (d, J1 ) 12 Hz, 1H); 31P (CD2Cl2), δ 18.8. IR (KBr pellet): 955 cm-1 (s, νRetO). UV-visible (C6H6): λmax/nm (log ), 579 (2.65), 428 (3.39), 370 (3.58). Anal. Found (calcd) for C31H25S3POClRe: C, 48.97 (48.84); H, 3.91 (3.31); S, 11.85 (12.62). [C2H5SReO(mtp)(PPh3)] (6b). This compound was prepared in 48% yield by the reaction of 6a and 1.0 equiv of LiSC2H5 (obtained as a green solid from C2H5SH and BuLi) in a procedure analogous to the method for [X-p-C6H4SReO(mtp)(PPh3)]. NMR: 1H (CD2Cl2), δ 7.747.69 (m, 6H), 7.54-7.44 (m, 9H), 7.42-7.41 (m, 1H), 7.37-7.33 (m, 1H), 7.30-7.28 (m, 1H), 7.19-7.16 (m, 1H), 4.92 (dd, J2 ) 12 Hz, J1 ) 1.6 Hz, 1H), 3.89-3.82 (m, 1H), 3.73-3.64 (m, 1H), 3.35 (d, J1 ) 12 Hz, 1H), 1.45 (t, J ) 7.2 Hz, 3H); 31P (CD2Cl2), δ 18.5. IR (KBr pellet): 950 cm-1 (s, νRetO). UV-visible (C6H6): λmax/nm (log ), 595 (2.57), 426(sh) (3.40), 392 (3.48), 350 (3.51). Anal. Found (calcd) for C27H26S3PORe: C, 48.64 (47.70); H, 4.30 (3.85); S, 13.60 (14.15). [o-Ph2PC6H4SReO(mtp)] (7). This compound was prepared in 82% yield with Li[o-SC6H4P(Ph)2] in a procedure analogous to the method for [C2H5SReO(mtp)(PPh3)]. NMR: 1H (CD2Cl2), δ 8.01-7.99 (m, 1H), 7.89-7.84 (m, 2H), 7.58-7.47 (m, 7H), 7.43-7.38 (m, 2H), 7.347.29 (m, 2H), 7.25-7.17 (m, 4H), 5.07 (dd, J1 ) 1.2 Hz, J2 ) 12 Hz, 1H), 3.25 (d, J1 ) 12 Hz, 1H); 31P (CD2Cl2), δ 55.9. IR (KBr pellet): 966 cm-1 (s, νRetO). UV-visible (C6H6): λmax/nm (log ), 565 (2.54), 448 (sh) (3.04), 380 (2.53), 344 (3.92). Anal. Found (calcd) for C25H20S3PORe‚0.5C4H10O: C, 48.18 (47.21); H, 3.57 (3.69); S, 13.75 (14.00). [PhOReO(mtp)(PPh3)] (6c). A solution of 6a (222 mg, 0.34 mmol) in THF (60 mL) was dropwise added to a solution of LiOPh (0.34 mmol) (from PhOH and BuLi), the resulting solution was stirred overnight, and then the solvents were removed in vacuo. The residue was dissolved in benzene (50 mL) and then filtered. The filtration was reduced to dryness and the obtained solid washed with diethyl ether to give the product (61 mg, 25%). NMR: 1H (CD2Cl2), δ 7.56-7.48 (m, 8H), 7.40-7.33 (m, 10H), 7.25-7.23 (m, 2H), 7.12-7.08 (m, 2H), 7.06-7.02 (m, 2H), 4.77 (d, J ) 12 Hz, 1H), 3.37 (d, J ) 12 Hz, 1H); 31 P (CD2Cl2), δ 32.5. IR (KBr pellet): 949 cm-1 (s, νRetO). UV-visible (C6H6): λmax/nm (log ), 557 (2.74), 458 (3.07), 352 (3.62). Analytically pure solids could not be isolated by column chromatography due to decomposition. J. AM. CHEM. SOC.
9
VOL. 127, NO. 29, 2005 10445
Li et al.
ARTICLES [ClReO(edt)(PPh3)] (5a). The procedure was as described for 6a: 1.0 g of [Cl3ReO(PPh3)2] (1.2 mmol) and 101 µL of 1,2-ethanedithiol (H2edt, 1.2 mmol) gave, after being stirred for 30 min and worked up similarly, 632 mg of a green solid (89%). NMR: 1H (CD2Cl2), δ 7.717.66 (m, 6H), 7.55-7.46 (m, 9H), 3.87-3.79 (m, 2H), 3.03-2.97 (m, 1H), 2.78-2.69 (m, 1H); 31P (CD2Cl2), δ 27.3. IR (KBr pellet): 978 cm-1 (s, νRetO). UV-visible (C6H6): λmax/nm (log ), 581 (1.97), 415 (3.42), 328 (3.48). Anal. Found (calcd) for C20H19S2POClRe: C, 40.10 (40.57); H, 3.33 (3.23); S, 10.02 (10.83). [C2H5SReO(edt)(PPh3)] (5c). A similar procedure was employed to obtain this product in 66% yield as described for [C2H5SReO(mtp)(PPh3)]. NMR: 1H (CD2Cl2), δ 7.75-7.71 (m, 6H), 7.54-7.44 (m, 9H), 3.94-3.92 (m, 1H), 3.82-3.73 (m, 1H), 3.65 (q, J ) 7.6 Hz, 2H), 2.64-2.50 (m, 2H), 1.38 (t, J ) 7.6 Hz, 3H); 31P (CD2Cl2), δ 20.8. IR (KBr pellet): 952 cm-1 (s, νRetO). UV-visible (C6H6): λmax/ nm (log ), 642 (1.61), 524 (2.44), 396 (3.60). Anal. Found (calcd) for C22H24S3PORe‚0.5H2O: C, 41.75 (42.15); H, 4.23 (4.02); S, 15.44 (15.35). [ReS(SC6H4CH2)(SC6H4-p-X))(PPh3)] (X ) OMe, Me, F, Cl) (2be). These complexes were all generated in procedures analogous to that described for 2a. We had no attempt to isolate the products to be fully characterized in pure form. The1H and 31P NMR spectra of the products are quite similar to those observed for 2a.2 [ReS(SC6H4CH2)(SC2H5)(PPh3)] (8). A glass flask was charged with [C2H5SReO(mtp)(PPh3)] (147 mg, 0.22 mmol), benzene (100 mL), and PPh3 (284 mg, 1.08 mmol). The mixture was stirred at ambient temperature for 5 days as the brown solution developed gradually. The solvent was reduced to dryness, and the residue was dissolved in a minimal volume of diethyl ether, loaded on a silica gel column, and eluted with hexanes-ether. The red-brown band was collected, and upon removal of the solvents a brown solid was obtained (20 mg, 14%). NMR: 1H (CD2Cl2), δ 7.79 (br, 6H), 7.61 (br, 2H), 7.52-7.45 (m, 9H), 7.12 (br, 2H), 6.67 (d, J ) 16.4 Hz, 1H), 4.09 (dd, J1 ) 16.4 Hz, J2 ) 9.6 Hz, 1H), 3.56 (q, J ) 6.8 Hz, 2H), 1.55 (t, J ) 6.8 Hz, 3H); 31 P NMR (CD2Cl2), δ 35.1. IR (KBr pellet): 521 cm-1 (s, νRetS). UVvisible (C6H6): λmax/nm (log ), 611 (1.68), 488 (3.16), 371 (4.04), 320 (3.93). Anal. Found (calcd) for C27H26S3PRe: C, 48.21 (48.85); H, 4.60 (3.95); S, 13.62 (14.49). [PhSReS(-SCHMeCH2CH2-)(PPh3)] (4b) and [PhSReS(SCH2CH2CHMe-)(PPh3)] (4c). A mixture of {PhSReO(pdtMe)}2 (241.7 mg, 0.28 mmol) and PPh3 (347 mg, 1.32 mmol) was dissolved in toluene (70 mL). The color of the solution rapidly changed from brown to yellowish as the reaction mixture was stirred at ambient temperature for 5 days and then filtered. The filtrate was evaporated to give a residue, which was dissolved in a minimal volume of diethyl ether. Column chromatography afforded a yellowish band, which was collected by eluting with ether-hexane; following that, the solvents were removed to give a solid. The pure product was obtained by washing with cool ether (yield 266 mg, 70%). IR (KBr pellet): 524 cm-1 (s, νRetS). UV-visible (C6H6): λmax/nm (log ), 594 (1.59), 458 (2.86), 386 (sh, 3.48), 337 (sh, 3.97). 324 (3.99). Anal. Found (calcd) for C28H28S3PRe: C, 49.63 (49.61); H, 4.60 (4.16); S, 14.08 (14.19). 4b. NMR: 1H (C6D6), δ 8.05 (br, 6H), 7.89-7.85 (m, 2H), 7.187.14 (m, 2H), 7.06-6.96 (m, 10H), 5.68 (br, d, J ) 14 Hz, 1Ha), 3.423.31 (m, 1Hb), 3.22-3.17 (m, 1Hc), 2.79-2.71(m, 1He), 1.79-1.68 (m, 1Hd), 1.55 (d, J ) 6.8 Hz, CH3); 13C (C6D6), δ 151.63, 151.52, 135.70, 135.60, 135.01, 134.49, 133.73, 133.69, 131.22, 131.20, 129.06, 128.94, 128.84, 127.32, 59.43 (d, CHe), 58.72 (d, CHaHb), 50.38 (CHcHd), 21.78 (CH3); 31P (C6D6), δ 31.8.
4c. NMR: 1H (C6D6), δ 8.05 (br, 6H), 7.89-7.85 (m, 2H), 7.187.14 (m, 2H), 7.06-6.96 (m, 10H), 5.62 (br d, 1Ha), 3.87-3.77 (m, 1Hb), 3.74-3.67 (m, 1He), 2.79-2.71 (m, 1Hc), 2.31-2.20 (m, 1Hd), 0.96 (d, J ) 6.8 Hz, CH3); 13C, δ 151.52, 151.39, 135.70, 135.60, 134.78, 134.27, 133.73, 133.69, 131.22, 131.20, 129.10, 128.89, 128.78, 127.27, 59.63 (d, CHe), 57.56 (d, CHaHb), 48.50 (CHcH d), 20.60 (CH3); 31 P (C6D6), δ 32.3.
{PhS(ReO)(pdtMe2)}2. 2,4-Pentanedithiol (103.2 mg, 0.76 mmol) was added by syringe to a suspension of {(ReO)(SPh)3}2 (401 mg, 0.0.38 mmol) in toluene (90 mL), and the mixture was stirred overnight under N2 and then filtered. The red-brown filtrate was evaporated to dryness and dissolved in CH2Cl2 (∼2 mL), which was loaded on silica gel. The first brown band was collected. Upon removal of the solvent, the obtained solid was washed with ether and vacuum-dried. Yield: 70 mg, 20%. 1H NMR (complicated owing to diastereomers) (CD2Cl2): δ 7.48-7.33 (m, 8H), 7.29-7.24 (m, 2H), 4.40-4.38 (m, 1H), 4.15-4.07 (m, 1H), 3.59-3.46 (m, 1H), 3.33-3.26 (m, 1H), 3.133.08 (m, 1H), 2.99-2.83 (m, 1H), 1.81-1.77 (m, 6H), 1.42-1.34 (m, 3H), 1.22-1.20, 1.21 (dd, J ) 6.8 Hz, 3H). Anal. Found (calcd) for C22H30O2S6Re2: C, 29.77 (29.65); H, 3.58 (3.39); S, 20.83 (21.59). [ReS(-SCHMeCH2CHMe-)(SPh)(PPh3)] (4d). Reaction of {PhS(ReO)(pdtMe2)}2 with excess PPh3 in C6D6 (∼0.8 mL) was monitored by 1H and 31P NMR spectroscopy at 25 °C. NMR: 1H (C6D6) (aryl region, overlapping with PPh3 and OPPh3), δ 6.11-6.05 (m, 1H), 3.973.93 (m, 1H), 2.61-2.54 (m, 1H), 2.40-2.33 (m, 1H), 2.41 (d, J ) 6.8 Hz, 3H), 1.21 (d, J ) 6.8 Hz, 3H); 31P (C6D6), δ 24.9. Kinetics Measurements. Reactions of the monomers with the phosphines were monitored by UV-visible spectra with an Agilent 8453 spectrophotometer with a pseudo-first-order excess of phosphine in benzene solution at 25.0 ( 0.2 °C. Initial monomeric Re(V) complex concentrations were [Re(V)]0 ) 0.08-0.23 mmol L-1 in systems containing 10-8000 equiv of phosphine. As the reactions proceeded, the solution color changed from green to red-brown in the mtp series, or from green to yellowish in the pdt series. Data analysis is described in the Results. X-ray Structure Determinations. The structures of the complexes in Table 4 were determined by X-ray crystallography. Selected crystals were mounted on a Siemens (Bruker) SMART 1000 CCD diffractometer equipped with a low-temperature apparatus. Mo KR radiation (λ ) 0.71073 Å) was used for data collection. Data were collected using the full sphere routine with 0.3° scans in ω with an exposure time 10-30 s/frame. Final cell parameters were retrieved by using SMART software and refined by using SAINT software, which corrects for Lorentz polarization and decay; absorption corrections were applied with SADABS.27 The space groups for all the complexes were assigned unambiguously by analysis of symmetry and systematic absences determined by the program XPREP. The structures were solved by direct or Patterson methods. The positions of the heavy atoms were found directly by both methods; the remaining atoms were located in an alternating series of least-squares cycles and difference Fourier maps. All non-hydrogen atoms were refined in full-matrix anisotropic approximation; all hydrogen atoms were placed in the structure factor calculation at idealized positions and were allowed to ride on the neighboring atoms with relative isotropic displacement coefficients. (27) All software and sources of the scattering factors are contained in the SHELXTL (version 5.1) program library (Sheldrick, G. M., Bruker Analytical X-ray Systems, Madison, WI).
10446 J. AM. CHEM. SOC.
9
VOL. 127, NO. 29, 2005
C−S Bond Cleavage Reactions of Re(V) Dithiolates
Acknowledgment. This research was supported by the U.S. Department of Energy, Office of Basic Energy Science, Division of Chemical Sciences, under Contract W-7405-Eng-82 with Iowa State University. We thank Adam J. Bergren for assistance with the electrochemistry and Steven Veysey for the elemental analyses. Supporting Information Available: 2D COSY spectrum and 1H
NMR spectra of the isomers 4b and 4c in C6D6 at 25 °C
ARTICLES
(Figures S1 and S2); linear correlations of kobs vs triarylphosphines concentration (Figures S3-S7); plot of log k2 vs numbers of the methyl groups on the pdt framework (Figure S8); cyclic voltammograms of the complexes with ReVO core (Figures S9S15) (PDF); complete crystallographic data for 1a, 5a, 6a, 6b, 7, and 4b (CIF). This material is available free of charge via the Internet at http://pubs.acs.org. JA0519334
J. AM. CHEM. SOC.
9
VOL. 127, NO. 29, 2005 10447